App-ID - Palo Alto Networks

advertisement
APP-ID
A foundation for visibility and control in
the Palo Alto Networks Security Platform
App-ID uses multiple identification techniques to
­determine the exact identity of applications traversing
your network – irrespective of port, protocol, evasive
tactic, or encryption. Identifying the application is the
very first task performed by App-ID, providing you with
the knowledge and flexibility needed to safely enable
applications and secure your organization.
App-ID™ is a patented traffic
classification technology that
identifies applications traversing
the network, irrespective of
port, protocol, evasive tactic or
encryption (TLS/SSL or SSH).
• Facilitates a more complete
understanding of the business
value and associated risk of
the applications traversing the
network.
• Enables creation and enforcement
of safe application enablement
policies.
• Brings application visibility and
control back to the firewall,
where it belongs.
Palo Alto Networks | Technology Brief
As the foundational element of our
enterprise security platform, App-ID
provides visibility and control over
applications – even those that try to
evade detection by masquerading as
legitimate traffic, hopping ports or
sneaking through the firewall using
encryption (TLS/SSL or SSH).
In the past, unapproved or non-workrelated applications on your network
left you with two choices—either
block everything in the interest of data
security, or enable everything in the
interest of business. These choices left
little room for compromise.
App-ID enables you to see the applications on your network and learn how
they work, their behavioral characteristics, and their relative risk. When used
in conjunction with User-ID™, you can
see exactly who is using the application
based on their identity, not just an IP
address. Armed with this information,
your security team can use positive
security model rules to allow the
applications that enable the business,
controlling them as needed to improve
your security posture.
Firewall Traffic Classification:
Applications, not Ports
Stateful inspection, the basis for most
of today’s firewalls, was created at
a time when applications could be
controlled using ports and source/
destination IPs. The strict adherence to
port-based classification and control is
foundational and cannot be turned off.
Even when augmented by “after the
fact” classifiers, applications cannot be
effectively controlled.
Palo Alto Networks® recognized that
applications had evolved to where
they can easily slip through the firewall
and chose to develop App-ID, an
innovative firewall traffic classification
technique that does not rely on any
one single element like port or protocol
to identify applications. Instead,
App-ID uses multiple mechanisms to
determine what the application is. The
application’s identity then becomes
the basis for firewall policy. App-ID has
been created to be highly extensible
and, as applications continue to evolve,
application detection mechanisms
can be added to App-ID or updated
1
Decryption
(SSL or SSH)
Decode
Check
Signatures
Policy
Check
K NO W N P R O T O C O L DEC ODE R
Check Application
Signatures
Policy Check
Check
IP/Port
Policy Check
Start
IDE N T IF IE D T R A F F IC (NO DEC ODING)
Apply Heuristics
Policy
Check
UNK NO W N P R O T O C O L DEC ODE R
REPORT & ENFORCE POLICY
Figure 1: How App-ID classifies traffic.
as a means of keeping pace with the
ever-changing application landscape.
App-ID Traffic Classification
Technology
in place, then decryption is not employed. A similar approach is used with
SSH to determine if port forwarding is
in use as a means to tunnel traffic over
SSH. Such tunneled traffic is identified
as ssh-tunnel and can be controlled via
security policy.
here, to provide visibility into applications that might otherwise elude positive identification. The actual heuristics
used are specific to an application and
include checks based on such things
as the packet length, session rate, and
packet source.
Using as many as four different
techniques, App-ID determines what
the application is as soon as the traffic • Application and Protocol Decoding:
With App-ID as the foundational
hits the firewall appliance, irrespective
element of our security platform, your
Decoders for known protocols are
of port, protocol, encryption (TLS/SSL
security team can regain visibility into,
used to apply additional context-based
or SSH) or other evasive tactic. The
and control over, the applications
signatures to detect other applications
number and order of identification
traversing your network.
that may be tunneling inside of the
mechanisms used to identify the
protocol (e.g., Yahoo! Messenger used
application will vary depending on
across HTTP). Decoders validate that
App-ID: Dealing with Custom or
the application. The general flow for
the traffic conforms to the protocol
Unknown Applications
App-ID is as follows:
specification and they provide supNew applications are added to the
port for NAT traversal and opening
App-ID database weekly, yet nearly
• Application Signatures: Signatures are
dynamic pinholes for applications such
every network will still have cases
used first to look for unique applias VoIP or FTP. Decoders for popular
where unknown application traffic
cation properties, and related transapplications are used to identify the
is detected. There are typically three
action characteristics, to correctly
individual functions within the appliscenarios where unknown traffic will
identify the application regardless of
cation as well (e.g., webex-file-sharbe detected: a commercially available
the protocol and port being used. The
ing). In addition to identifying applicaapplication unknown to App-ID, an
signature also determines if the applitions, decoders identify files and other
internal custom application, or a threat.
cation is being used on its default port
content that should be scanned for
or a non-standard port (for example,
• Unknown Commercial Applications:
threats or sensitive data.
RDP across port 80 instead of port
Using visibility tools, you can quickly
3389, its standard port). If the identi- • Heuristics: In certain cases, evasive
determine if the traffic is a commercial
fied application is allowed by security
applications still cannot be detected
off-the-shelf (COTS) application. If it is
policy, further analysis of the traffic is
even through advanced signature and
a COTS application, you can capture
done to identify more granular appliprotocol analysis. In those situations, it is
and submit traffic packets to Palo Alto
cations as well as scan for threats.
necessary to apply additional heuristic,
Networks for App-ID development.
or behavioral analysis to identify cerThe new App-ID is developed, tested,
• TLS/SSL and SSH Decryption: If App-ID
tain applications, such as peer-to-peer
and added to the database for all users
determines that TLS/SSL encryption
file sharing or VoIP applications that
in the form of a weekly update.
is in use and a decryption policy is
use proprietary encryption. Heuristic
in place, the traffic is decrypted and
• Internal or Custom Applications: If the
analysis is used as needed, with the
then passed to other identification
application is internal, or custom, you
other App-ID techniques discussed
mechanisms as needed. If no policy is
Palo Alto Networks | Technology Brief
2
can create a custom App-ID using a
set of available protocol and application decoders. Once the custom
App-ID is developed, your internal application is classified and inspected in
the same manner as applications with
standard App-IDs. Custom App-IDs
are managed in a separate database
on the device, ensuring they are not
impacted by the weekly (commercial)
App-ID updates.
• Threats: Once the commercial and
internal applications have been
addressed, the third possible source
of unknown traffic is threats. Here
too, you can quickly determine the
risk l­evels using the behavioral botnet
­report or other forensics tools to
isolate the characteristics and apply
appropriate policy control.
Even after attempts to identify, some
traffic in the system may remain
unknown. Because our firewall supports
a positive enforcement model, the
remaining unknown traffic can be
blocked (by default) or allowed but
tightly controlled by policy if desired.
Alternative offerings (e.g., Intrusion Prevention Systems) are based on negative
control and will allow unknown traffic
to pass through without providing any
semblance of visibility or control.
How App-ID Works: Identifying
WebEx
When a user initiates a WebEx session,
the initial connection is an encrypted
communication. With App-ID, the
device sees the traffic and the signatures determine that it is using TLS/SSL.
The decryption engine and protocol
decoders are then initiated to decrypt
the TLS/SSL and detect that it is HTTP
traffic. Once the decoder has the HTTP
stream, App-ID can apply contextual
signatures and detect that the application in use is WebEx. At this point
the session traffic becomes known as
WebEx traffic by the firewall. Visibility
(e.g., ACC in the user interface) and
control of the WebEx traffic via security
policy are enabled.
traversing your network. Learning
what the application does, the ports it
uses, its underlying technology, and its
behavioral characteristics is the next
step toward making a more informed
decision about how to treat the application. Once a complete picture of usage
is gained, you can apply policies with a
range of responses. Examples include:
• Allow or deny
• Allow but scan for exploits, viruses
and other threats
• Allow based on schedule, users or
groups
• Control file or sensitive data transfer
• Decrypt and inspect
• Apply traffic shaping through QoS
If the end user were to initiate the
• Apply policy-based forwarding
WebEx Desktop Sharing feature, this
“mode-shift” from conferencing to
• Allow a subset of application
remote access would be detected by
­functions
App-ID. Again, visibility to this specific
application function would be ­provided – • Any combination of the above
and policy control over WebEx ­Desktop
With App-ID as the foundational
Sharing would be possible (distinct from
element of our firewalls, you can
general WebEx use).
restore visibility and control over the
applications and traffic traversing your
Application Identity: The Heart of
network.
Policy Control
Identifying the application is the first
step in learning more about the traffic
Figure 2: Application Function Control – maximize productivity by safely enabling the
application itself (Microsoft SharePoint) or individual functions.
Palo Alto Networks | Technology Brief
3
Application Function-Level Controls
To many customers, safe application
enablement means striking an appropriate security policy balance by enabling
some application functions while
blocking others. Examples include:
• Allowing Microsoft® SharePoint®
Documents, but blocking the use of
SharePoint Administration.
• Block Facebook®-mail, -chat, -posting
and -apps, but allow Facebook itself,
effectively only allowing users to
browse Facebook.
• Enable the use of MSN®, but disable the
use of MSN-file transfer – or only allow
certain file types to be transferred.
Using an application hierarchy that
includes the base application and
supporting functions, App-ID makes it
easy for you to choose which applications to allow overall, while blocking
or controlling functions within the
application. Figure 2 shows SharePoint as the base application, and the
individual functions within.
Controlling Multiple Applications:
Dynamic Filters and Groups
Figure 3 below shows a snapshot view
of Palo Alto Networks online application database. Here you can browse the
current database of App-IDs, including
an interactive view of applications
based on the same criteria that can be
used in dynamic filters.
There are cases where you may want
to control larger groups of applications
in bulk, as opposed to controlling them
individually. The two mechanisms that
address this policy requirement are
dynamic filters and application groups. • Application Groups: An application
group is defined as a static list of
Dynamic Filters: A dynamic filter is a
applications. An example would be a
set of applications that is created based
group of remote management applion any combination of the filter criteria:
cations such as RDP, Telnet, and SSH.
category, subcategory, behavioral charIn a typical organizational scenario,
acteristic, underlying technology or risk
each of these applications is used by
factor. Security policies (e.g. deny, allow,
support and IT personnel, yet employscan) can be applied to dynamic filters. ees who fall outside of these groups
The security policy is then enforced
are also known to use them as a means
for application traffic that matches the
to access their home networks. An apfilter criteria.
plication group can be created, with an
As new App-IDs are introduced and
associated security policy that allows
delivered to the firewall via weekly
use only by support and IT personupdates, dynamic filters are automatnel (supported by User-ID). As new
ically updated for those applications
employees join the organization, they
that meet the filter criteria. This helps
need only be added to the appropriate
minimize administrative effort associatdirectory group. No updates are needed with security policy management.
ed to the security policy itself.
Figure 3: Browse up-to-date application research and analysis at the Palo Alto Networks
Application Research Center, https://applipedia.paloaltonetworks.com
Palo Alto Networks | Technology Brief
4
Expanding the List of Applications
The list of App-IDs is expanded weekly
with 3 to 5 new applications typically
added based on input from customers,
partners, and market trends. When you
find unidentified applications on your
network, you can capture the traffic and
then submit the information for App-ID
development. Once a new App-ID is
­ eveloped and tested, it is added to
d
the list as part of the weekly content
updates.
Summary
App-ID is a powerful and differentiated core capability of the Palo
Alto N
­ etworks Next-Generation
Firewall, enabling advanced visibility
4401 Great America Parkway
Santa Clara, CA 95054
Main:+1.408.753.4000
Sales:+1.866.320.4788
Support:+1.866.898.9087
www.paloaltonetworks.com
and granular control of traffic in your
­network. With this visibility and control,
you can evaluate what is taking place
in your environment, and then define
policies that: ensure appropriate use,
reduce ­attack surface, and stop threats.
Ultimately, App-ID is a key foundational
element in enabling superior risk
management for your organization.
© 2015 Palo Alto Networks, Inc. Palo Alto Networks is a registered trademark
of Palo Alto Networks. A list of our trademarks can be found at http://www.
paloaltonetworks.com/company/trademarks.html. All other marks mentioned
herein may be trademarks of their respective companies.
pan-tb-app-id-092115
Download